714 



Fishery Bulletin 91(4), 1993 



were found among viable, non-viable and stillborn 

 pups. 



A simple linear regression model defined significant 

 latitudinal variation in birth timing of viable pups with 

 a gradient of 4.10 days/°latitude; this slope was not 

 significantly different from that reported by Temte et 

 al. (1991) for colonies of wild Pacific harbor seals be- 

 tween 30° and 47°N. A second order polynomial model 

 (Fig. 5; Table 2), however, was more statistically ap- 

 propriate based on residual analysis. This model was 

 highly significant (r 2 =0.508; F l2107l =55.23; P<0.0001), 

 defining a positive relationship between latitude and 

 birth date. 



Data regarding sex of pup (n = 102), maternal age 

 (n=88), and the previous annual cycle of the mother 

 (pregnant vs. non-pregnant: rc=89) were available for 

 subsets of the harbor seals. When entered, either sepa- 

 rately or as a group, into a multiple regression model 

 adjusting for latitude, none of these parameters was 

 found to significantly alter birth timing. 



Discussion 



Use of captive birth data 



Events on rookery sites or pupping beaches hamper 

 the estimation of true pupping seasons in pinnipeds. 

 For example, premature pupping in the California sea 

 lion occurs with increasing frequency over a 5-month 

 period, melting into the normal pupping season be- 

 tween mid-May and the end of June (Delong et al., 

 1973). In addition, seasonal movements of animals from 

 breeding to feeding sites may occur (Braham, 1974; 

 Mate, 1975), potentially obscuring latitudinal varia- 

 tion of birthing if seasonal entrainment occurs at lati- 



Pacific Harbor Sea! 



Latitude (°N) 



Figure 5 



The birth timing of viable Pacific harbor seals as a function 

 of latitude. See Table 2 for regression model. 



tudes other than that of the rookery site. The use of 

 captive populations permits accurate measurement of 

 birth dates and birth outcomes, while controlling for 

 potential latitudinal displacement during the repro- 

 ductive year. 



California sea lion 



This study confirms the report of Schusterman et al. 

 (1982) and demonstrates that marked latitudinal varia- 

 tion occurs not only in the timing of birth, but also in 

 the variance of the mean birth date. Furthermore, in- 

 terpretations of results based on this data set repre- 

 senting 466 viable births from 41 locations, as com- 

 pared to the three locations used by Schusterman et 

 al. (1982), are far less sensitive to possible confound- 

 ing effects induced by captivity or differences in cli- 

 matic conditions. The extreme example of sea lions 

 translocated to the Southern Hemisphere provides fur- 

 ther evidence of strong latitudinal effect. 



The birth timing of captives mirrors that of wild 

 populations. Although a review by Mate (1979) con- 

 cluded that pups, regardless of latitude, are born from 

 mid-May to late-June over the range of California sea 

 lion rookeries, slight latitudinal variation may exist. 

 In Baja California, pupping occurs in late-June 

 (Brownell et al., 1974), and Le Boeuf et al. (1983) re- 

 ported a maximum pup count on 10 July at Los Islotes 

 (28°N). Further to the north, at San Nicolas Island 

 (33°15'N), pupping peaks during the first half of June 

 (Peterson and Bartholomew, 1967; Odell, 1975; Heath 

 and Francis 1 -). Data from more northerly colonies are 

 anecdotal in nature. For example, Braham (1974) re- 

 ported a 2-week-old pup on 11 June at San Luis Obispo 

 County, California (35°30'N). 



As in captive sea lions, premature pupping in wild 

 colonies occurred as early as January on San Nicolas 

 and San Miguel Islands (Odell, 1970; Delong et al, 

 1973). Premature pups are probably represented in 

 captivity by the stillborn and non-viable groups. The 

 209r rate of stillborn and non-viable pupping for cap- 

 tive sea lions is noteworthy considering the 5-16 f /< rate 

 of premature pupping reported for San Nicolas Island 

 during years with high incidences of prematurity 

 (Odell, 1970). Whereas pesticide exposure and disease 



'Heath. C, and J. Francis. 1983. California sea lion population dy- 

 namics and feeding ecology with applications for management. Re- 

 sults of 1981-1982 research on Santa Barbara and San Nicholas 

 Islands. U.S. Dep. Commer., NOAA, Natl. Mar. Fish. Serv., South- 

 west Fish. Sci. Center. P.O. Box 271. La Jolla, CA 92038. Admin. 

 Rep. LJ-83-04C. 



Heath, C, and J. Francis. 1984. Results of research on California 

 sea lions, San Nicholas Island, 1983. U.S. Dep. Commer., NOAA. 

 Natl. Mar. Fish. Serv., Southwest Fish. Sci. Center, P.O. Box 271, La 

 Jolla, CA 92038. Admin. Rep. LJ-84-41C. 



